This presentation will explore data gathering techniques, tools, and analysis processes in the business innovation process. By way of example, the presentation will outline the stages of planning, designing, and delivering behind one of today’s most popular business innovation use cases for IoT – a predictive maintenance system. It will also reveal the different areas in which businesses gain value (and cost savings) by automating the processes of data engineering and data discovery.

1. Big Data Driven
Business Innovation
Wael Elrifai
Director of Enterprise Solutions

2. Innovating in Transport
Business Challenges
 Modernize and improve rail transportation reliability
in the U.K
 Reduce maintenance costs
Use Case
 XaaS / Usage based pricing
 Predictive Maintenance (PdM) & Schedule
Optimisation, Total Asset Optimisation
 Internet of Things
 Big Data Science

3. Cs
of
IoT
Connections
Conversions
Centralisation
Cognition
Continuous improvement
What Is IoT?

4. Data Science Techniques
Unsupervised
Learning
Supervised
Learning
Reinforcement
Learning
Computational
Statistics &
Mathematical
Optimisation
Statistics & Machine
Learning

5. Statistical Models: Ordinary Least Squares

6.  Supervised learning
‒ Discover patterns in the data that relate data attributes with a target attribute.
‒ Use these patterns to predict the values of the target attribute in future data instances.
 Unsupervised learning
‒ No target attributes.
‒ We want to explore the data to find some intrinsic structures.
Supervised vs. Unsupervised Learning

7. Machine Learning: Supervised Learning

8. Machine Learning: Unsupervised Learning
K – Means Clustering

9. Ask yourself the following question…
Where Do I Start?
“What could I do if I had perfect,
universal, and timely information?”

10. Thank You